AMERICAN JOURNAL OF HUMAN BIOLOGY 21:2–15 (2009)

Feature Article

Epigenetics and the Embodiment of Race: Developmental Origins of

US Racial Disparities in Cardiovascular Health
CHRISTOPHER W. KUZAWA* AND ELIZABETH SWEET
Department of Anthropology, Northwestern University, Evanston, Illinois 60208

ABSTRACT The relative contribution of genetic and environmental influences to the US black-white disparity in
cardiovascular disease (CVD) is hotly debated within the public health, anthropology, and medical communities. In this
article, we review evidence for developmental and epigenetic pathways linking early life environments with CVD, and
critically evaluate their possible role in the origins of these racial health disparities. African Americans not only suffer
from a disproportionate burden of CVD relative to whites, but also have higher rates of the perinatal health disparities
now known to be the antecedents of these conditions. There is extensive evidence for a social origin to prematurity and
low birth weight in African Americans, reflecting pathways such as the effects of discrimination on maternal stress
physiology. In light of the inverse relationship between birth weight and adult CVD, there is now a strong rationale to
consider developmental and epigenetic mechanisms as links between early life environmental factors like maternal
stress during pregnancy and adult race-based health disparities in diseases like hypertension, diabetes, stroke, and cor-
onary heart disease. The model outlined here builds upon social constructivist perspectives to highlight an important
set of mechanisms by which social influences can become embodied, having durable and even transgenerational influen-
ces on the most pressing US health disparities. We conclude that environmentally responsive phenotypic plasticity, in
combination with the better-studied acute and chronic effects of social-environmental exposures, provides a more parsi-
monious explanation than genetics for the persistence of CVD disparities between members of socially imposed racial
categories. Am. J. Hum. Biol. 21:2–15, 2009. ' 2008 Wiley-Liss, Inc.

The disproportionate disease and mortality burden of Tang et al., 2005), and that disease-causing alleles are
African Americans is among the most challenging of US likely to be among those variants that segregate between
public health problems. It is now broadly known that an these groups (Burchard et al., 2003; Risch et al., 2002).
African American man in Harlem is less likely than a man Evidence to support this model has recently come from
in Bangladesh to survive to the age of 65 (McCord and genetic studies of population substructure, in which the
Freeman, 1990). Nationally, African Americans have an analysis of thousands of loci simultaneously has produced
age-adjusted all-cause mortality rate that is 1.5 times that clusters of genetic information that can be used to cor-
of whites (Keppel et al., 2002), and cardiovascular dis- rectly identify individuals’ self-described geographic
eases (CVDs) and their precursor conditions, including ancestry (Redon et al., 2006; Rosenberg et al., 2002; Tang
hypertension, diabetes, and obesity, contribute heavily to et al., 2005).
this disparity. The risk of dying from heart disease is 1.3 Those who argue that social forces drive racial health
times higher in African Americans compared to US whites disparities point to the importance of factors such as eco-
(Mensah et al., 2005), and African Americans are 1.8 nomic disadvantage, psychosocial stress, and institutional
times more likely to develop diabetes (CDC, 2007). Hyper- and interpersonal discrimination as causes of ill health
tension rates are roughly 1.5–2 times higher in African (Brondolo et al., 2003; Davidson et al., 2000; Dressler,
Americans compared to whites (Mensah et al., 2005), and 1991; Harrell et al., 2003; Jonas and Lando, 2000; Sweet
are especially high in certain regions, such as the so-called et al., 2007; Troxel et al., 2003; Williams, 1999; Williams
stroke belt of the American South. In total, nearly half of and Collins, 1995; Williams and Jackson, 2005; Williams
all African American adults develop some form of CVD, and Neighbors, 2001; Wyatt et al., 2003). Such cultural
making racial disparities in these conditions one of the and structural challenges can impose barriers to healthy
most pressing US public health problems today (AHA, lifestyles, limit access to quality medical care, and chroni-
2007). cally strain physiological stress systems that are linked to
During the past 15 years, there has been a concerted disease (Dressler et al., 2005; Kaplan and Lynch, 2001;
effort to understand the underlying determinants of racial Krieger, 2005; Krieger and Davey Smith, 2004; LaVeist,
health disparities (Krieger, 2005; Lillie-Blanton and Lave- 2005; McEwen, 2001). Together, these social, economic,
ist, 1996; Williams, 1999), and explanations have tended and contextual factors can have a significant impact on
to align with one of two models that emphasize either health, and when taken into account, health disparities
social or genetic causes. Researchers who attribute some,
or all, of the problem of racial health inequalities to differ- *Correspondence to: Christopher Kuzawa, Department of Anthropology,
ences in genetic predisposition (Burchard et al., 2003; Northwestern University, 1810 Hinman Avenue, Evanston, IL 60208,
Hirsch et al., 2006; Risch et al., 2002; Sarich and Miele, USA. E-mail: kuzawa@northwestern.edu
Received 20 May 2008; Revision received 17 July 2008; Accepted 29 July
2004; Saunders, 1995) assume that human genetic varia- 2008
tion can be differentiated into conventional racial clusters DOI 10.1002/ajhb.20822
(Allocco et al., 2007; Calafell, 2003; Hinds et al., 2005; Lao Published online 16 October 2008 in Wiley InterScience (www.interscience.
et al., 2006; Redon et al., 2006; Rosenberg et al., 2002; wiley.com).

C 2008
V Wiley-Liss, Inc.
 EPIGENETICS AND THE EMBODIMENT OF RACE 3
between African Americans and US whites are often developmental mechanisms that underlie these associa-
diminished (Dressler et al., 2005; McDade et al., 2006; tions is increasingly being pursued under the rubric of the
Williams and Collins, 1995). ‘‘developmental origins of health and disease’’ (DOHaD)
The debate between these competing models has been (see Gluckman and Hanson, 2006), while the public health
described as a ‘‘storm’’ (Krieger, 2005), and mirrors long- impacts of early life influences is an important focus of the
standing discussions in anthropology over the meaning of burgeoning field of ‘‘life course epidemiology’’ (see Kuh
the race concept (Armelagos and Goodman, 1998). As with and Shlomo, 2004; Smith, 2003).
early theories of racial-genetic determinism, current Long-term impacts of early life undernutrition or stress
genetic models of racial health disparities have been have been proposed to help explain patterns of adult CVD
criticized on several fronts, including the sampling biases risk in a variety of ecological, political economic, and cul-
that have been present in studies of population substruc- tural settings. To date, much emphasis has been given to
ture (Serre and Paabo, 2004), and the low percentage of the potential role of these processes in transitional popu-
genetic variation that is typically explained by ‘‘racial’’ lations in which a combination of poor early life nutrition
clusters (e.g.  4% Rosenberg et al., 2002). This latter followed by adult weight gain in the same generation
criticism is supported by over three decades of research could lead to elevated CVD risk (Adair and Prentice, 2004;
consistently showing between-group genetic differences to Benyshek et al., 2001; Gluckman and Hanson, 2005;
be small compared to the genetic variation found within Kuzawa and Adair, 2003; Prentice and Moore, 2005). Sim-
continental regions (Brown and Armelagos, 2001; Good- ilarly, the tendency for overweight and diabetic women to
man, 2000; Jorde and Wooding, 2004; Kittles and Weiss, give birth to overweight, diabetes-prone offspring, operat-
2003; Lewontin, 1972; Relethford, 2002). Despite this ing through intrauterine influences on developmental
evidence and the demonstrated importance of social envi- pathways, has been proposed as an explanation for
ronmental factors for African American health, the tend- the high rates of diabetes and metabolic syndrome in
ency for self-identified race to remain a significant predic- populations in South Asia and the American Southwest
tor of disease outcomes in epidemiological studies, even af- (Benyshek et al., 2001; Yajnik, 2004).
ter lifestyle and SES factors have been adjusted for Although a variety of life course models of cardiovascu-
statistically (Cooper, 1993; Otten et al., 1990; Pappas lar epidemiology have been developed (Ben-Shlomo and
et al., 1993), continues to be interpreted by some as indi- Kuh, 2002), there has been little systematic evaluation of
rect support for the racial genetic position (Kistka et al., the potential contribution of developmental and epigenetic
2007). responses to early environments to the specific problem of
As many scholars of health inequality have observed, US black-white health disparities in CVD. There is good
however, the impact of the social environment on health is reason to expect that the DOHaD field will help clarify the
multifaceted and challenging to adequately measure and origin of these racial health disparities (see also Kuzawa,
adjust for statistically (Braveman et al., 2005; Kaufman 2008). African Americans not only have higher rates of
and Cooper, 1999; Smith, 2000). One dimension of this CVD as adults, but they also have a higher burden of the
problem is the often low resolution of conventional mea- antecedent condition of lower birth weight—an early life
sures of social, economic, and behavioral determinants of health disparity believed to trace in part to factors like
health. More nuanced approaches to quantifying stress stress and discrimination experienced by the mother dur-
and other social, cultural, and material processes related ing pregnancy and across her life course (Pike, 2005).
to cardiovascular health are needed, and recent work in Thus, there is a strong rationale to consider a develop-
biocultural anthropology has made critical contributions mental and transgenerational dimension to these racial
to this area of health disparities research (Dressler and disparities in cardiovascular health.
Bindon, 2000; Gravlee et al., 2005). This article does not comprehensively review the
In addition, there is a growing appreciation that envi- DOHaD or life-course epidemiology literatures, which
ronmental influences contribute to adult health dispar- have been the subject of extensive recent reviews (Ben-
ities by influencing biological processes and responses Shlomo and Kuh, 2002; Gluckman and Hanson, 2006;
across the life cycle, with certain ages or developmental Kuh et al., 2003; Smith, 2003). Nor is this paper intended
stages particularly sensitive to environmental and social as a comprehensive review of the causes of racial health
influence (Barker, 1994). Building from earlier studies disparities or of the full breadth of life-course influences
showing that adult mortality is predicted by socioeco- on adult health (for more see Geronimus, 2001; Krieger,
nomic conditions experienced around the time of birth 2000; Pollitt et al., 2005; Williams, 2005). Rather, our goal
(Forsdhal, 1977; Kermack et al., 1934), research during is to evaluate the potential contribution of one specific set
the past two decades has now established that early life of biological pathways to the problem of cardiovascular
conditions, such as prenatal undernutrition and stress, or health disparities between African Americans and US
maternal stress during pregnancy, can modify develop- whites: the influence of maternal health and stress during
mental biology in offspring in a fashion that elevates their pregnancy on the development of fetal biological systems,
risk of developing diseases like diabetes, hypertension, which can elevate CVD risk in adult offspring. To this
and CVD as adults (Barker and Osmond, 1986; Gluckman end, we first briefly review evidence for maternal-fetal
et al., 2008). Fields including clinical and animal model influences on systems that influence adult CVD risk, and
research, epigenetics, anthropology, public health, sociol- discuss the role of developmental and epigenetic processes
ogy, and economics are addressing the evolutionary ori- as underlying mechanisms. Next, we discuss evidence
gins of these developmental responses, their biological that these pathways are likely operative in African Ameri-
bases, and their health and policy implications (e.g. cans, and that they help explain US racial disparities in
Forrest and Riley, 2004; Geronimus et al., 2006; Graham adult CVD. We argue that the embodiment of social and
and Power, 2004; Halfon and Hochstein, 2002; Kuzawa material environments through developmental and epige-
and Pike, 2005; Palloni, 2006). Study of the biological and netic processes helps explain the persistence of biological

American Journal of Human Biology

4 C.W. KUZAWA AND E. SWEET

CVD disparities across racial categories that are socially health characteristics with birth weight data recorded in
rather than genetically defined (Krieger, 2005). birth records and largely ignored other aspects of the social
Controversy remains over the terminology used in environment, such as socioeconomic status, that might
scholarly research to describe human biological variation, account for the associations (Kramer and Joseph, 1996;
and a consensus is lacking within anthropology (Gravlee Paneth et al., 1996). Nearly two decades of research have
and Sweet, 2008). Given lack of supportive evidence that helped push the field beyond this initial skepticism, and
race is a genetically meaningful concept, some scholars DOHaD is now a well-established area of study that lies at
have opted to use terms such as ‘‘ethnicity’’ or ‘‘population’’ the intersection of fields like medicine, public health, and
to describe geographically or culturally identified groups, anthropology (Gluckman and Hanson, 2006; Kuzawa and
while others continue to use the term ‘‘race’’ when refer- Pike, 2005). Hundreds of human studies have replicated
ring to the social phenomenon of historically constructed findings of developmental programming, many incorporat-
racial categories (AAA, 1998; di Leonardo, 2004; Harrison, ing longitudinal data on a range of lifestyle and environ-
1995; Shanklin, 1994; Weismantel, 1997). Although rele- mental influences that might confound associations with
vant for understanding disparities in disease, the concept birth size (Adair et al., 2001; Dalziel et al., 2007; Gupta
of ‘‘ethnicity’’ traditionally includes a broad set of cultural et al., 2007; Huxley et al., 2007; Law et al., 2001; Levitt
practices and shared beliefs that define group identity et al., 2000; Miura et al., 2001; Tian et al., 2006). These
(Gordon, 1964). In this article, we choose to use the term studies find that smaller birth size predicts higher blood
‘‘race’’ because many of the social forces we discuss as pressure (reviewed by Adair and Dahly, 2005), insulin re-
underlying determinants of health disparities, such as dis- sistance and diabetes (Eriksson et al., 2002; Yajnik, 2004),
crimination, economic inequalities, or segregated neigh- abnormal cholesterol profiles (Kuzawa and Adair, 2003),
borhoods, represent the unique lived reality of race as a an ‘‘android’’ or abdominal pattern of fat deposition (Oken
socially-defined and imposed system in the US. In light of and Gillman, 2003), and an elevated risk of suffering or
the lack of consensus surrounding terminology, we empha- dying from CVD (Huxley et al., 2007; Leon et al., 1998).
size that we define race as a socially constructed category Conditions experienced during infancy and childhood have
that has biological implications, rather than a genetically also been shown to predict adult biological and health out-
justified criteria for classifying human variation (AAA, comes. Not unlike birth size, small size in infancy is also
1998; Cooper and David, 1986). associated with higher CVD risk in adulthood, while
breastfed infants have lower rates of hypertension, obesity,
BACKGROUND and diabetes as adults (Arenz et al., 2004; Lawlor et al.,
Early environments and adult health 2005). There is also evidence that prenatal and postnatal
exposures interact to influence adult health. For instance,
For the past two decades, evidence has been accumulat- being born small but then experiencing rapid weight gain
ing that stress, prenatal nutrition, and other early life fac- during childhood—a marker of later nutritional abun-
tors can influence risk for adult cardiovascular and meta- dance—predicts the same constellation of adult diseases
bolic diseases. Starting in the late 1980s, David Barker (Adair and Cole, 2003; Ong, 2006).
and colleagues at Southampton University published a se- Because birth weight reflects both environmental and
ries of papers showing that the risk of dying from CVD, or genetic factors, a relationship between birth weight and
of suffering from conditions that precede CVD like hyper- adult physiology or disease risk could simply reflect the
tension or diabetes, is higher among individuals who pleiotropic effects of genes. As one example, insulin regu-
weighed less at birth (Barker, 1994; Barker and Osmond, lates fetal growth but also has broad involvement in adult
1986; Barker et al., 1989). Although studies had previ- metabolic disease, and pleiotropic genes that influence in-
ously found evidence for relationships between depriva- sulin metabolism could yield a correlation between fetal
tion during childhood and higher subsequent adult mor- growth and conditions like insulin resistance or diabetes
tality rates (Forsdahl, 1977; Kermack et al., 1934), the as a result of genetic rather than developmental processes
Southampton group was the first to link these associations (Freathy et al., 2007; Hattersley and Tooke, 1999).
to a biological marker that hinted at possible mechanisms Although birth weight is fraught with interpretive chal-
to account for them. lenges (Kuzawa and Adair, 2004), multiple observations
Building from the assumption that a baby born small demonstrate that genetic correlations do not fully explain
had been poorly nourished prior to birth, they proposed the associations documented between birth weight and
that these relationships were the outcome of adjustments CVD risk in humans.
made by the fetus in response to a compromised intrauter- The first line of evidence is the generally low heritabil-
ine nutritional environment. They reasoned that a fetus ity of birth weight. Although high heritabilities for birth
faced with undernutrition would not only slow its growth weight are occasionally reported for studies in well-nour-
rate to reduce nutritional requirements, but might also ished pedigrees (e.g. US/Fels: 0.82, Demerath et al., 2007;
modify the structure and function of organs and systems 0.59 broad sense, Stern et al., 2000), most studies find
involved with metabolism and physiology, with effects that genetic inheritance accounts for only a fraction of the
that could linger into adulthood to influence risk of devel- variance in birth weight. Based upon twin registries, her-
oping chronic disease. Such durable alterations to devel- itabilities for birth weight are typically reported in the
opmental biology in response to early environments have range 0.2–0.4 (e.g. (Baird et al., 2001; Vlietinck et al.,
been described as developmental ‘‘programming’’ (Dörner, 1989; Whitfield et al., 2001), with national birth weight
1975; Lucas, 1991) or ‘‘induction’’ (Bateson, 2001). registry studies finding similar estimates (0.31 for birth
The hypothesis that adult metabolism, biology, and dis- weight and 0.27 for birth length in all Norwegian births
ease risk could be ‘‘programmed’’ by prenatal nutrition was from 1967 to 2004; Lunde et al., 2007). The remaining var-
greeted with skepticism (Kramer and Joseph, 1996; iance is believed to be determined by maternal influences
Paneth et al., 1996). Most early studies merely linked adult like nutritional status, exposure to stress, or other factors

American Journal of Human Biology

 EPIGENETICS AND THE EMBODIMENT OF RACE 5
influencing blood flow to the endometrial lining or pla- undernourished individuals tend to be smaller and to
centa (Gluckman and Hanson, 2004). have fewer nephrons, making them more prone to hyper-
Second, in studies of monozygotic twin pairs, the lighter tension and renal failure later in life (Lampl et al., 2002;
twin has been shown to have elevated risk for adverse Luyckx and Brenner, 2005). Similarly, alterations in the
changes in body composition and risk for diabetes and number and composition of muscle cells in individuals
hypertension later in life (Bo et al., 2000; IJzerman et al., born lower birth weight could contribute to insulin resist-
2003; Iliadou et al., 2004), demonstrating that genetic cor- ance in adulthood (Jensen et al., 2007), as could changes
relations do not fully account for the associations with in the type and number of body fat cells present in differ-
adult disease risk. Finally, observational research on ent adipose depots (Zhang et al., 2007).
humans is broadly corroborated by animal model In addition to such modifications in the number of cells
research. Animal work in the DOHaD literature has con- present, there is growing evidence that epigenetic changes
firmed that factors influencing prenatal nutrition and in the pattern of cellular gene expression are also key to
intrauterine environmental conditions can induce physio- the long-term impacts of early environments (Sinclair
logic and metabolic changes in offspring that linger into et al., 2007; Waterland and Michels, 2007). Although
adulthood (Sinclair et al., 2007; Symonds and Gardner, ascribed with numerous meanings since Waddington
2006). For instance, restricting the nutritional intake of coined the phrase in 1942 (Waddington, 1942), epigenetics
pregnant rats, mice, or sheep, or directly restricting blood is increasingly being reserved to refer to the study of proc-
flow (and thus nutrient transfer) to the fetus, increases esses that modify patterns of gene expression without
postnatal blood pressure, cholesterol, abdominal fat depo- changing the nucleotide sequences of the DNA (Jenuwein
sition, and diabetes risk in offspring (reviewed by Lang- and Allis, 2001). The genome is inherited at conception
ley-Evans et al., 2003; McMillen and Robinson, 2005). and, other than somatic mutations acquired during cell di-
Although the wide availability of birth weight data has vision, remains unchanged in most body cells across the
led to a research emphasis on the role of nutrition as a lifecycle. The ‘‘epigenome,’’ in contrast, is the product of a
programming stimulus, maternal psychological stress gradual commitment of cell lineages to more constrained
during pregnancy can lead to a similar constellation of bi- patterns of gene expression. The epigenome is a result, in
ological changes and disease risk factors in adult off- part, of the genome interacting with the environment, and
spring, and at times independent of changes in birth can be viewed as the molecular basis for cellular differen-
weight. The fetus is normally shielded from exposure to tiation and development over the life-course (Fig. 1).
stress hormones produced by the mother’s body by an Unlike the nucleotide bases that form the genetic code,
enzyme (11-bHSD) that is expressed by the placenta the ‘‘epigenetic code’’ predominantly involves chemical
where it converts the active form of the hormone (cortisol modifications to the structure of the chromatin that scaf-
in humans) to its inactive form (cortisone). This buffering folds the DNA within the chromosomes (Berger, 2007). If
capacity can be exceeded when the mother is severely fully stretched, the chromosomes in a single human cell
stressed, leading to premature or excessive exposure of would be roughly 6 feet in length; thus, a complex process
the fetus to maternal stress hormones. This in turn can of folding is required to package the complete genome into
contribute to reduced birth size by directly reducing fetal each cell nucleus where the genes reside and are
growth rate. Although the pathways are not fully under- expressed. In the nucleus, chromosomes must be
stood, it can also influence the stress hormone-related cas- unwound locally to allow transcription factors to gain
cade that triggers parturition, leading to early pregnancy access to a gene. How the DNA is packaged within the
termination (Challis et al., 2005). This fetal exposure to chromatin influences how easy or difficult a gene is to
excess cortisol induces a similar suite of biological changes access and thus, whether and how much it may be
in offspring as are observed with dietary restriction, expressed in that cell. Epigenetic markings have thus
including an elevation in blood pressure, stress reactivity, been likened to volume controls for genes, and they play
abdominal adiposity, insulin resistance, and other precur- an integral role in the normal process of cellular differen-
sors of diabetes and CVD (Seckl and Meaney, 2004). Thus, tiation. As cells divide, epigenetic markings present in the
prenatal stress—whether nutritional or psychosocial in parent cell are maintained through mitosis and thus her-
origin—shapes a wide range of traits that influence future itable to both daughter cells (but see Suzuki and Bird,
risk of developing CVD, including how the body manages 2008). Through a complex series of bifurcations at which
and distributes glucose and lipids, regulates blood pres- patterns of gene silencing and amplification are progres-
sure, and responds physiologically to stress. sively acquired, the single totipotent ‘‘stem cell’’ formed at
conception is capable of creating a body with roughly 200
The mechanisms of phenotypic ‘‘memory’’—growth, cell types that vary in structure and function, despite the
development, and the epigenetic code endowment of each of these daughter cells with an identi-
cal genome (Reik, 2007).
The durability of the effects of early environments on An important class of mechanisms of epigenetic gene
multiple biological systems raises the question of what bi- silencing involves localized chemical modifications to the
ological mechanisms underlie them: if early environments chromatin and its protein constituents, which alter how
influence adult biology and health, where in the body are tightly the DNA is packaged in the region of specific
the ‘‘memories’’ of these early experiences stored and genes. The attachment of an extra methyl group (methyla-
maintained? The contributions of several developmental tion) to ‘‘CpG islands’’ (regions of DNA rich in cytosine
processes have been documented, each corresponding to and guanine linked by a phosphodiester bond) within the
axes of biological variation independent of one’s genotype. promoter region of a gene typically impedes expression of
The most straightforward involves a change in growth of a that gene in that cell (Berger, 2007). The histone proteins
tissue or organ as reflected in its size or cell number. As that the DNA fibers are wrapped around can also be modi-
one well-documented example, the kidneys of prenatally fied to alter the tightness of DNA packing, and thus the

American Journal of Human Biology

6 C.W. KUZAWA AND E. SWEET

Fig. 1. Schematic illustrating the role of epigenetic gene silencing in the differentiation of an initially totipotent stem cell (the zygote) to ‘‘com-
mitted’’ daughter cell lineages. Gray horizontal lines indicate genes capable of being transcribed to produce a protein, whereas black lines are
genes that have been silenced by epigenetic modifications (for simplicity, processes that enhance gene expression, such as histone acetylation,
are not shown). The pattern of gene silencing is heritable to daughter cells, leading to the eventual commitment of cell lineages to specialized cell
types (e.g. neurons, muscle cells) as epigenetic marks are accumulated. The focus of classical genetics on modeling the determinants and evolu-
tionary change in gene frequencies is concerned with the genes inherited at conception (the genome), while epigenetics focuses on the narrower
pattern of gene silencing and expression in the cells of specific tissues, organs, and systems (the epigenome). Although epigenetic changes in
gene expression are largely regulated themselves by genes, environmental exposures can modify some epigenetic marks in specific cells lines
during growth and development, which partly accounts for the durable effects that early environments have on adult biology and disease risk.

accessibility of that stretch of DNA to enzymes and tran- netic marks and downstream patterns of gene expression
scription factors. Methylation of the histone generally in specific cells and cell lineages (Gluckman et al.,
impedes gene expression, whereas acetylation loosens the 2007a,b; Ho and Tang, 2007; Jirtle and Skinner, 2007;
chromatin and promotes gene expression. Although more Waterland and Jirtle, 2004; Waterland and Michels,
commonly implicated in cancers than CVDs, another 2007). Recent experimental studies in animal models dem-
epigenetic mechanism involves small noncoding RNA onstrate how epigenetic markings in offspring may
(‘‘micro RNA’’ or ‘‘small RNA’’) (Grewal and Elgin, 2007) respond to maternal factors like diet (Lillycrop et al.,
which are produced in the cell nucleus. Although not tran- 2005) and rearing behavior (Weaver et al., 2004). In preg-
scribed to make proteins themselves, they block transcrip- nant rats, protein restriction during gestation reduces
tion and expression of other genes in a gene-specific fash- methylation of the promoter region of the gene that codes
ion (RNA interference or ‘‘RNAi’’), thus providing another for the glucocorticoid receptor (GR)—the receptor that rec-
way that gene expression can be modified in a durable ognizes and responds to the stress hormone cortisol (a glu-
fashion. cocorticoid)—in offspring liver cells. Because methylation
impedes access of transcription factors to the gene’s pro-
Epigenetics and adult cardiovascular disease risk moter region, the reduced methylation triggered by this
dietary intervention increases expression of the GR gene,
Current research is showing how environmental factors thus increasing the number of receptors expressed in the
can modify epigenetic processes, thereby affecting epige- liver. This results in an amplification of the liver’s meta-

American Journal of Human Biology

 EPIGENETICS AND THE EMBODIMENT OF RACE 7
bolic response to stress hormones, for example increasing The emerging understanding of the epigenetic mecha-
expression of the downstream gene product PEPCK—the nisms that build the phenotype represents a revolution in
rate-limiting enzyme in glucose production (gluconeogen- biology that is gathering momentum (Baylin and Schue-
esis) (Lillycrop et al., 2007). In this particular animal bel, 2007; Kennedy, 2002). Processes such as promoter
model, the nutritional experiences of one generation dur- region methylation reveal why knowing an organism’s ge-
ing pregnancy (the pregnant rat dam) influence how the notype—the genes inherited by the totipotent zygote at
offspring regulate and produce glucose in response to conception—is merely the first frame in the story of how
stress as adults. Maternal protein restriction has also the phenotype is eventually built. Attempts to identify
been shown to induce hypomethylation of the angiotensi- susceptibility genes for diseases involving complex sys-
nogen receptor gene in the adrenal gland in rat offspring, tems and traits have generally had poor results (Cooper
which likely helps explain why the gene is overexpressed, and Psaty, 2003). By demonstrating one important way
contributing to the elevated systolic blood pressure seen that the impact of a gene on the phenotype can be modi-
in these animals (Bogdarina et al., 2007). fied by the environment, this new understanding of epige-
In other instances, the effect of early environments can netic processes is helping shed light on this issue. By link-
linger beyond adulthood to be passed on to future genera- ing maternal experience with fetal biology, this literature
tions. Such examples of transgenerational epigenetic in- is showing how stressors experienced by one generation,
heritance can occur through several types of pathways such as imbalanced nutrition or psychosocial stress, can
(Gluckman et al., 2007a; Jablonka and Lamb, 1995). perpetuate changed biological settings to offspring, with
Although less often studied, research suggests a more lim- effects on such functions as glucose metabolism, blood
ited but biologically important lingering impact of pater- pressure regulation, fat deposition, and the physiologic
nal stress experience on the biological characteristics of response to stress.
offspring (e.g. Drake et al., 2005). These, and related find-
ings in females, suggest that some environmentally AN EPIGENETIC MODEL OF BLACK-WHITE
induced cytoplasmic elements or epigenetic markings are DISPARITIES IN CVD
already present in egg or sperm at conception (Anway
et al., 2005; Cropley et al., 2006; Stöger, 2008). In females, The adult CVDs in which epigenetic and developmental
provisioning of metabolic and other resources via placenta processes play a critical role are the very ones that exhibit
and lactation opens up additional pathways by which epi- the most pronounced disparities across racial groups. As
genetic settings can be perpetuated across generations. we now review, the following observations suggest that
When the establishment of epigenetic markings in off- epigenetic and developmental responses contribute to
spring cells is sensitive to environmental exposures dur- race-based US health disparities: (1) As is true for a wide
ing the period of direct dependence—i.e. when the ‘‘envi- range of human populations, birth outcomes are impor-
ronment’’ is the maternal phenotype—this can recapitu- tant predictors of adult cardiovascular health for African
late a pattern of epigenetic marks in offspring in the Americans; (2) African American mothers have higher
absence of direct transfer of those marks through sperm rates of low-birth-weight births than white mothers in the
or egg (Drake and Walker, 2004). US; (3) this racial disparity in birth outcomes is linked to
One well-documented example of such epigenetic reca- environmental, and particularly psychosocial, factors, and
pitulation is anxiety and rearing style in lactating rats (4) there is evidence that these patterns can have multi-
(Diorio and Meaney, 2007; Weaver et al., 2004). Rat pups generational consequences.
reared by indulgent mothers exhibit changes in methyla-
tion of the GR gene in hippocampal neurons involved in Birth weight and adult CVD risk in African Americans
regulating the hypothalamic-pituitary-adrenal (HPA) axis
and the stress response. This has the effect of reducing While few large studies have been conducted among
reactivity and anxiety in offspring, and encourages them diverse US populations, past research shows that the
to adopt a more relaxed and indulgent rearing approach effects of prenatal environments on African American
with their own offspring (the grandoffspring). Cross-fos- health are generally in agreement with expectations from
tering of pups shows that this effect is not genetic but de- other populations. Several small US studies have shown
pendent upon maternal rearing behavior. The effect is that lower birth weight predicts higher blood pressure,
also reversed by chemically blocking epigenetic marking, elevated cortisol reactivity, and early signs of diabetes in
showing that it is not simply a learned behavior (Weaver older African American children and adolescents (Covelli,
et al., 2004). This study illustrates how a maternal pheno- 2006a,b; Li et al., 2001, 2006; Oberg et al., 2007), as well
type can construct a rearing environment that tends to as other related cardiovascular conditions, such as end-
replicate the same phenotype in the next generation, oper- stage renal disease, in adults (Fan et al., 2000). Findings
ating not through genes or learning, but through trans- from larger, population-based cohort studies have demon-
generational recapitulation of epigenetic marking. Indi- strated the most consistent evidence for the effects of birth
rect evidence for maternal-fetal transfer of epigenetically weight on subsequent health among African Americans.
based alterations in stress (HPA) reactivity has been docu- In the well-characterized Bogalusa Heart Study, birth
mented in humans. In holocaust survivors, severity of weight is inversely related to later systolic and diastolic
post traumatic stress disorder (PTSD) symptoms—which blood pressure in adult African Americans (Donker et al.,
influences maternal cortisol production during preg- 1997; Mzayek et al., 2007). Biracial analyses from that
nancy—predict levels of cortisol excretion in postnatal off- study suggest that for some cardiovascular risk factors,
spring, and Manhattan women who were pregnant during such as blood pressure, cholesterol levels, and insulin
the 9/11 attacks gave birth to offspring who show evidence resistance, birth weight may be a stronger predictor for
for alterations in HPA activity in childhood (Yehuda and African Americans than for whites (Donker et al., 1997;
Bierer, 2008). Mzayek et al., 2004). Thus, as for other US and global pop-

American Journal of Human Biology

8 C.W. KUZAWA AND E. SWEET

ulations, and consistent with experimental findings in have all been linked with higher risk for LBW deliveries.
animal models, lower birth weights predict elevated Racial discrimination in particular has been shown to con-
future adult risk for adverse cardiovascular outcomes in fer a twofold or higher increased risk for poor birth out-
African Americans. comes (Collins et al., 2004; Dole et al., 2003; Mustillo
et al., 2004), and in one study that pooled a multiracial
sample this accounted for a substantial portion of the
Lower African American birth weight observed racial difference in preterm deliveries (Mustillo
et al., 2004). Together these findings suggest that social
It is well established that African Americans have lower
factors, especially those relating to the experience of
average birth weights than US whites. National data
stress and inequality, contribute to the lower average
show that rates of low birth weight (LBW) deliveries are
birth weights in African American pregnancies.
twice as high among African Americans compared to
Further evidence for an environmental, rather than
whites, and very LBW births (<1,500 g) are 2.69 times
genetic, cause of the lower birth weights of African Ameri-
more common among African Americans (CDC, 2005;
cans comes from studies of multigenerational trends of
Keppel et al., 2002). This pattern of racial disparity is true
birth outcomes. A nongenetic transgenerational influence
for both main categories of LBW: preterm (Demissie et al.,
on fetal growth has long been proposed in the medical
2001) and small for gestational age (SGA) births
community (Ounsted and Ounsted, 1968; Ounsted et al.,
(Alexander et al., 1999). The racial disparity in birth out-
1986). Maternal fetal growth rate is among the strongest
comes has been documented for several decades and has
predictors of offspring fetal growth rate (Morton, 2006;
shown no signs of significant improvement during that
Ramakrishnan et al., 1999), and among survivors of the
time (Demissie et al., 2001; Kramer et al., 2006).
Dutch Famine winter during WWII, the grandoffspring of
pregnant women who experienced the famine had reduced
Social origins of African American low birth weight fetal growth (Lumey, 1992). Given evidence for effects of
the mother’s early life and chronic experiences on the
The association of birth weight with adult CVDs lends intrauterine environment that she provides offspring,
urgency to the search for the causes of the lower average women of the same ‘‘race’’ might be expected to give birth
birth weights of African Americans compared to other de- to larger or smaller babies, depending on where they were
mographic subgroups in the US. As with attempts to born and raised. There is in fact good evidence for such
explain other health disparities, hypotheses have tended differences.
to align with either genetic or environmental explana- Many studies have compared the birth weights and
tions. While a genetic cause is a theoretical possibility, perinatal health of recent immigrants to the US (who
there is no evidence that genetic differences between were born overseas) to their racial or ethnic counterparts
groups explain these inequalities, and, as we discuss born in the US (Kleinman et al., 1991). These studies are
below, epidemiologic evidence is difficult to reconcile with remarkably consistent in their findings. African American
this interpretation (see also Pike, 2005). newborns in general have higher rates of LBW, PTD, and
Because maternal stressors and the passage of stress neonatal mortality relative to whites in the US. However,
hormones across the placenta can lead to both preterm these differences are greatly reduced among African
birth and fetal growth restriction (Sandman et al., 1997), American offspring born to foreign-born mothers. In one
research has examined the contribution of psychosocial study of nearly 2.5 million US deliveries, foreign-born
stress to LBW and preterm delivery (PTD) in African women of African ancestry were 25% less likely to give
Americans. Several epidemiologic studies have found that birth to a LBW baby compared to their US-born counter-
stressful life conditions and specific measures of psychoso- parts, while there was no difference in birth outcome by
cial stress are associated with increased risk for both pre- natality among whites (Acevedo-Garcia et al., 2005). Sev-
term birth and fetal growth restriction in African Ameri- eral other studies report similar findings, showing that
can mothers (Giscombe and Lobel, 2005). Exposure to foreign-born African Americans giving birth in the US
stressful life events among African American mothers is have rates of LBW that are closer to those of US whites
associated with a higher risk for preterm births and lower than US-born African Americans (Cabral et al., 1990;
birth weight (Borders et al., 2007; Dole et al., 2003; Domi- Forna et al., 2003; Singh and Yu, 1996).
nguez et al., 2005; Orr et al., 1996; Oths et al., 2001). One study of Illinois birth records not only compared
Additionally, psychological and emotional correlates of birth outcomes in foreign-born and US-born African
stress, such as symptoms of depression and anxiety, have Americans but also linked these data with information on
been linked with poorer birth outcomes for African Ameri- birth weights across several generations of offspring sub-
can women (Dole et al., 2003; Mackey and Boyle, 2000; sequently born in the US. The patterns present in the first
Orr et al., 1996). generation were similar to those described above: In con-
Several factors related to racial and economic inequality trast to the lower birth weights of US-born African Ameri-
in US society have also been found to predict adverse birth cans, foreign-born African Americans were found to have
outcomes. Factors related to socioeconomic status, such as a birth weight distribution nearly identical to that of US
income, education, and access to prenatal care, which whites (David and Collins, 1997). However, this equiva-
tend to be lower among African Americans, are related lence was short lived. Among subsequent generations
to birth outcomes for this population in some studies born in the US, the birth weight distribution of the off-
(Giscombe and Lobel, 2005; Wightkin et al., 2007). Expo- spring of African immigrants shifted to lower values (Fig.
sure to racial discrimination (Collins et al., 2004; Dole 2), en route to a convergence with the lower African Amer-
et al., 2004; Mustillo et al., 2004), residential segregation ican mean (Collins et al., 2002). The findings among the
(Bell et al., 2006; Grady, 2006), and neighborhood-level European immigrants in this study showed the opposite
poverty (Farley et al., 2006; Reagan and Salsberry, 2005) pattern: their birth weights were originally lower than

American Journal of Human Biology

 EPIGENETICS AND THE EMBODIMENT OF RACE 9

Fig. 2. Schematic illustrating the intergenerational change in birth

weight among recent African immigrants to the US. The first genera-
tion in the US, born to foreign-born mothers, has a mean birth weight
and birth weight distribution comparable to that of US whites. Second
and third generations born in the US have lower birth weights, mov-
ing closer to the African American mean. Not drawn to scale (after
data in Collins et al., 2002; David and Collins, 1997).

the mean for US whites, but increased with each genera-

tion born in the US.
It goes without saying that these opposing biological
responses were far too rapid to be due to changes in gene
frequencies (Boas, 1912). Instead they reveal that living
in the United States has different implications for the Fig. 3. Model showing the intergenerational transmission of dis-
intrauterine environments that African Americans and ease states operating through the reciprocal effects of a stressful
whites experience prior to birth as reflected in differences intrauterine environment on future adult metabolic state, and adult
in fetal growth rate, prematurity, and birth weight. metabolic state (in females) on a stressful intrauterine environment
in the next generation. The experience of chronic stress can have both
Regardless of where populations emigrate from, after acute and cumulative adverse effects on the present generation, and
several generations the birth weight distribution of later among women, lingering effects on future generations of offspring
generations comes to resemble that of their US ethnic operating through durable epigenetic changes (modified after Drake
counterparts. This convergence is strong evidence that AJ and Walker BR, 2004, Journal of Endocrinology, 180(1):1–16.
the widely documented US racial difference in birth
Ó Society for Endocrinology 2004. Reproduced with permission).
weight is not due to genes (Collins et al., 2002; David and
Collins, 2007). risk for having a preterm birth or LBW delivery by as
much as two to three times (Allen et al., 2004; Ananth
et al., 1995; Clausson et al., 1998; Graham et al., 2007;
Transgenerational impacts
McCowan et al., 1996; Ray et al., 2001). Similarly, mater-
In addition to the links between maternal stress and nal insulin resistance, hyperinsulinemia, and diabetes
the offspring’s future health, there are several pathways during pregnancy can increase passage of glucose and in-
through which the effects of a stressful intrauterine envi- sulin across the placenta, and encourage the development
ronment could be perpetuated across generations. The of a similar state of weight gain and metabolic dysregula-
most straight-forward explanation for a perpetuation of tion in offspring (Dabelea et al., 2000; Lampl and Jeanty,
risk involves a continuity of environments. Given the per- 2004; Pettitt et al., 1987, 1988; Silverman et al., 1995).
sistence of racial institutional discrimination and eco- Heightened stress reactivity not only restricts fetal
nomic inequality in US society (Blane et al., 1999; Pollitt growth and increases risk for premature delivery, but
et al., 2005; Shapiro, 2004; Williams and Collins, 1995), there is evidence that it can also have direct effects on the
LBW infants are likely to experience many of the same development of the fetus’s HPA axis (Worthman and
psychosocial stressors as adults that their parents did. Kuzara, 2005). As we discuss in greater detail below, these
Thus women who were themselves born small will likely epigenetic effects are not set in stone, and may be amen-
be at high social-environmental risk for delivering LBW able to reversal by intervention. However, these findings
offspring as a result of the perpetuation of a similar social suggest that the intrauterine environment experienced by
and economic environment. one generation (the mother) can influence the intrauter-
It is important to note, however, that the ‘‘environment’’ ine environment that she creates for her offspring, in
that a fetus experiences is an expression of maternal phe- theory helping perpetuate certain biological or metabolic
notype. This opens up possibilities for a mother’s own states, albeit in a fading fashion, across multiple matrilin-
stressful prenatal experience, as reflected in her having eal generations (Fig. 3).
been born small, to influence the intrauterine develop- The evidence reviewed above describes the components
mental environment she provides for the next generation. of a developmental model of US black-white disparities in
Hypertension during pregnancy, for instance, elevates cardiovascular health. Few US studies with high-quality

American Journal of Human Biology

10 C.W. KUZAWA AND E. SWEET

birth weight data have had sufficiently large samples are symptomatic of structural inequality and discrimina-
across demographic subgroups to empirically test the con- tion rather than choice. The most important predictors of
tribution of birth outcomes to adult racial health dispar- compromised birth outcomes include factors such as self-
ities. However, a recent analysis of data from the biracial perceived discrimination, racism, and chronic stress (Gis-
Bogalusa Heart Study cohort has provided strong support combe and Lobel, 2005; Mustillo et al., 2004). These expe-
for a developmental origin of a key racial health inequal- riences are no more the ‘‘choice" of the women who experi-
ity. In this study, the hypertension disparity between US ence them than are the many other symptoms of racial
whites and African Americans—one of the most common discrimination that have been documented in US society,
and widely studied racial health differential—was no lon- such as African Americans’ lower average incomes (Sha-
ger significant after models adjusted for the effects of piro, 2004) and reduced job opportunities compared to
birth weight (Cruickshank et al., 2005). This is one of the whites with equivalent qualifications (Pager, 2003).
rare studies to have ‘‘explained away’’ the race disparity The emerging epigenetic model of health disparities
in an adult CVD risk factor. No genetic factors have been points to social and economic change as key to addressing
shown to do this, despite considerable research effort racial differences in disease burden, and underscores the
(Cooper and Psaty, 2003). need to implement these interventions across the life-
course. In particular, this work opens up the possibility
DISCUSSION for new approaches to encouraging positive health states
in future generations. Some sources of social inequality,
The epigenetic and developmental processes that we such as racism, cannot be eliminated by legislation. But
review are shedding new light on the health disparities societies can legislate changes in public spending that
debate. In the current polarized discourse over health in- benefit pregnant mothers, improve their access to
equality, some interpret the inability of adult socioeco- adequate prenatal care and nutrition and help ensure
nomic and behavioral factors to account for racial dispar- that they are relatively buffered from stress while preg-
ities in disease burden as evidence for underlying genetic nant and lactating. Although evidence is mixed (Lu et al.,
differences (e.g. Kistka et al., 2007). This reasoning can be 2005), improving access to social support has been found
critiqued for ignoring the substantial residual impact of in some studies to reduce rates of LBW among African
social and environmental factors not captured in the lim- American women at high risk for adverse pregnancy out-
ited measures of these exposures employed in epidemio- comes (Norbeck et al., 1996). Promotion of breastfeeding,
logic research (Cooper et al., 2003). Evidence for develop- and longer and more secure maternity leave, are addi-
mental and epigenetic influences on adult health adds a tional examples of policies that could have long-term
new layer to this critique. As the research reviewed here health benefits for future generations, and ease race-
illustrates, measuring the biological impact of social forces based health differentials operating through developmen-
solely at the level of the adult phenotype misses important tal pathways.
developmental and epigenetic pathways that likely con- A better understanding of the epidemiology of epige-
tribute to racial health inequality. A genetic interpreta- netic processes will be critical in developing effective
tion of the residual race effect problematically conflates interventions (Waterland and Michels, 2007). Although
observed biological variation with inferred genetic contri- birth weight data are routinely collected in epidemiologic
butions, and ignores evidence that social factors can have research and are thus widely available for such studies,
durable life-course and transgenerational effects on birth weight is at best a nonspecific indicator of genetic,
health. Whereas group membership and continental race epigenetic, and other factors. Future research will benefit
are poor predictors of genetic variation, these same cate- from incorporation of more nuanced approaches to quanti-
gories are directly related to the social and structural fying stress and other social, cultural and material proc-
manifestations of inequality that impact the development esses that could influence the nutritional and endocrine
of responsive biological systems. A wealth of evidence now characteristics of the prenatal environment. For example,
shows that the social and economic experiences of race ethnographic approaches to the social and cultural con-
have profound influence on adult health and, beginning in texts of stress are providing improved insights into the
childhood, can have effects that are both chronic and cu- causes and impacts of stress in different communities and
mulative in their impact (Dressler et al., 2005; Geroni- demographic subgroups (Dressler and Bindon, 2000;
mus, 2001). The research reviewed here is bolstering this Gravlee et al., 2005), and will have much to add to future
social constructivist perspective by highlighting specific work on the developmental origins of adult health
developmental pathways through which these same social disparities.
factors become embodied during early, critical periods in While we have emphasized the role of the prenatal envi-
development, with impacts that extend into adulthood ronment in this review, the impact of stress, nutrition,
and at times even across generations (Krieger, 2005). and other social-environmental exposures on developmen-
Some may be tempted to interpret these findings as tal biology are by no means limited to fetal life. Infancy,
stigmatizing for pregnant women, or shifting blame onto childhood, and adolescence are all critical developmental
mothers for the long-term health consequences of stressful windows during which epigenetic modifications in gene
prenatal environments. The deleterious effects of some expression and tissue and organ function take place. As
maternal behaviors on offspring health, such as smoking mentioned earlier, there is evidence that breast feeding
or excessive drinking during pregnancy, have long been confers protection against developing obesity, diabetes,
appreciated (Leary et al., 2006), and indeed, the DOHaD and CVD (Arenz et al., 2004; Lawlor et al., 2005). The
literature broadens the scope of offspring health outcomes quality of the rearing environment and emotional attach-
that might be adversely affected by such behaviors (e.g. ment can have lasting effects on reactivity of the stress
Oken et al., 2008). However, the research reviewed here hormone (HPA) axis (Gunnar, 1998) and are influenced by
overwhelmingly points to the importance of factors that factors like maternal emotional well-being (Adam et al.,

American Journal of Human Biology

 EPIGENETICS AND THE EMBODIMENT OF RACE 11
2004). Rapid weight gain during childhood can increase The model presented here should not be understood as
adult risk for CVDs and exacerbate the impacts of a replacing genetic race with an essentialized concept of epi-
stressful prenatal environment, suggesting that interven- genetic race; instead, it shows how social environments,
tions to limit childhood weight gain could disproportion- defined along lines of constructed and socially imposed
ately benefit the health of individuals born small or racial identities, can drive developmental processes,
exposed to a stressful prenatal environment (Adair and thereby becoming embodied as biological patterns that
Cole, 2003; Oken and Gillman, 2003). Additionally, brain influence health and disease (Krieger, 2005). Debates
regions linked to emotional processing and stress reactiv- about the causes of racial health disparities have tradi-
ity, as well as other aspects of the HPA axis, undergo criti- tionally aligned with the poles in the classic model of dis-
cal structural development during adolescence (McCor- ease causation, which emphasizes the contrasting disease
mick and Mathews, 2007; Romeo and McEwen, 2006), impacts of inherited genes and the dynamic and culturally
suggesting that this age is also an important period for shaped environment. The emerging epigenetic and devel-
programming of the physiologic stress response. opmental model of chronic disease epidemiology illus-
In animal models, this flexibility of the phenotype dur- trates why this perspective is incomplete, and must be
ing later developmental periods has been found to allow broadened to account for the more durable role that envi-
for partial or complete reversal of some epigenetic ronments have on patterns of biology and health when
responses to prior stressful environments. Recent animal experienced early in the lifecycle.
research has shown that environmental enrichment dur-
ing adolescence can reverse some of the deleterious effects ACKNOWLEDGMENTS
of early life epigenetic programming (Bredy et al., 2003;
Francis et al., 2002; Laviola et al., 2004). Similarly, rat Yarrow Axford, Douglas Bolender, Richard Cooper,
models have shown that injection of neonates with leptin Thomas McDade, Dorothy Roberts, Zaneta Thayer, Robert
(a body-fat-derived hormone that signals energy status) Waterland, and two anonymous reviewers provided help-
reverses the adverse metabolic changes triggered by pre- ful feedback on this manuscript.
natal protein restriction (Gluckman et al., 2007b; Vickers
et al., 2005), while orally administered leptin in suckling LITERATURE CITED
rats (perhaps mimicking lactation in a well-nourished
AAA. 1998. American Anthropological Association Statement on ‘‘Race’’.
mother) protects the offspring against developing obesity Arlington, VA: American Anthropological Association.
later in life (Pico et al., 2007). These studies demonstrate Acevedo-Garcia D, Soobader MJ, Berkman LF. 2005. The differential effect
the continued flexibility of biological systems into later of foreign-born status on low birth weight by race/ethnicity and educa-
stages of development, and hold open the possibility that tion. Pediatrics 115:e20–e30.
Adair L, Dahly D. 2005. Developmental determinants of blood pressure in
strategies can be developed to modify disease risk and adults. Annu Rev Nutr 25:407–434.
reverse epigenetic influences established prior to birth. Adair LS, Cole TJ. 2003. Rapid child growth raises blood pressure in ado-
Thus, while the social consequences of race can have dura- lescent boys who were thin at birth. Hypertension 41:451–456.
ble effects on biology and health, we stress that ‘‘durable’’ Adair LS, Kuzawa CW, Borja J. 2001. Maternal energy stores and diet
composition during pregnancy program adolescent blood pressure. Cir-
need not equate with ‘‘permanent.’’ culation 104:1034–1039.
Adair LS, Prentice AM. 2004. A critical evaluation of the fetal origins hy-
pothesis and its implications for developing countries. J Nutr 134:191–
193.
Epigenetics as a challenge to the concept of genetic race Adam EK, Gunnar MR, Tanaka A. 2004. Adult attachment, parent emo-
tion, and observed parenting behavior: mediator and moderator models.
As emphasized by Boas (1912) a century ago, the contin- Child Dev 75:110–122.
gency of the adult phenotype on environmental conditions AHA. 2007. American Heart Association heart facts 2007: all Americans.
experienced during growth and development poses a fun- Dallas, TX: American Heart Association.
Alexander GR, Kogan MD, Himes JH, Mor JM, Goldenberg R. 1999. Racial
damental challenge to essentialist concepts of race. Cur- differences in birthweight for gestational age and infant mortality in
rent research on developmental and epigenetic contribu- extremely-low-risk US populations. Paediatr Perinat Epidemiol 13:205–
tions to adult health disparities is updating Boas’ argu- 217.
ment. Not only are traditional racial categories poor Allen VM, Joseph K, Murphy KE, Magee LA, Ohlsson A. 2004. The effect
of hypertensive disorders in pregnancy on small for gestational age and
predictors of gene frequencies, a fact that has been appre- stillbirth: a population based study. BMC Pregnancy Childbirth 4:17.
ciated for decades (Brown and Armelagos, 2001; Lewontin, Allocco DJ, Song Q, Gibbons GH, Ramoni MF, Kohane IS. 2007. Geography
1972), but developmental and epigenetic processes help to and genography: prediction of continental origin using randomly
selected single nucleotide polymorphisms. BMC Genomics 8:68.
clarify why genes do not determine biological fates in any Ananth CV, Peedicayil A, Savitz DA. 1995. Effect of hypertensive diseases
simple fashion. Genes rarely ‘‘determine’’ phenotypes but in pregnancy on birthweight, gestational duration, and small-for-gesta-
instead set the range of outcomes that a biological system tional-age births. Epidemiology 6:391–395.
may create as it interacts with and responds to the devel- Anway MD, Cupp AS, Uzumcu M, Skinner MK. 2005. Epigenetic transge-
nerational actions of endocrine disruptors and male fertility. Science
opmental environment (Bogin, 1999; Kuzawa, 2005; West- 308:1466–1469.
Eberhard, 2003; Worthman and Kuzara, 2005). Humans Arenz S, Ruckerl R, Koletzko B, von Kries R. 2004. Breast-feeding and
inhabit highly variable and socially stratified ecologies; it childhood obesity—A systematic review. Int J Obes Relat Metab Disord
follows that systems that coordinate adaptation to these 28:1247–1256.
Armelagos G, Goodman A. 1998. Race, racism, and anthropology. In:
realities should come equipped with a capacity to organize Goodman A, Leatherman T, editors. Building a new biocultural synthe-
in response to local patterns of stress and opportunity sis: political-economic perspectives on human biology. Ann Arbor: Uni-
(Kuzawa, 2008). The research that we review is demon- versity of Michigan Press. p 359–377.
strating some of the pathways through which socially Baird J, Osmond C, MacGregor A, Snieder H, Hales CN, Phillips DI. 2001.
Testing the fetal origins hypothesis in twins: The Birmingham Twin
defined environmental context can become embodied, con- Study. Diabetologia 44:33–39.
tributing to the local perpetuation of linked patterns of Barker D. 1994. Mothers, babies, and disease in later life. London: BMJ
early life and adult health disparity. Publishing.

American Journal of Human Biology

12 C.W. KUZAWA AND E. SWEET

Barker DJ, Osmond C. 1986. Infant mortality, childhood nutrition, and Covelli M. 2006a. The relationship of low birth weight to blood pressure,
ischaemic heart disease in England and Wales. Lancet 1:1077–1081. cortisol levels, and reactivity in African American adolescents: a pilot
Barker DJ, Osmond C, Golding J, Kuh D, Wadsworth ME. 1989. Growth in study. Issues Compr Pediatr Nurs 29:173–187.
utero, blood pressure in childhood and adult life, and mortality from Covelli MM. 2006b. The relationship of blood pressure and cortisol reactiv-
cardiovascular disease. BMJ 298:564–567. ity to family history of hypertension of African American adolescents. J
Bateson P. 2001. Fetal experience and good adult design. Int J Epidemiol Cardiovasc Nurs 21:347–353.
30:928–934. Cropley JE, Suter CM, Beckman KB, Martin DI. 2006. Germ-line epige-
Baylin S, Schuebel K. 2007. The epigenomic era opens. Nature 448:548– netic modification of the murine A vy allele by nutritional supplementa-
549. tion. Proc Natl Acad Sci USA 103:17308–17312.
Bell JF, Zimmerman FJ, Almgren GR, Mayer JD, Huebner CE. 2006. Birth Cruickshank JK, Mzayek F, Liu L, Kieltyka L, Sherwin R, Webber LS, Sri-
outcomes among urban African-American women: a multilevel analysis navasan SR, Berenson GS. 2005. Origins of the ‘‘black/white’’ difference
of the role of racial residential segregation. Soc Sci Med 63:3030–3045. in blood pressure: roles of birth weight, postnatal growth, early blood
Ben-Shlomo Y, Kuh D. 2002. A life course approach to chronic disease epi- pressure, and adolescent body size: The Bogalusa Heart Study. Circula-
demiology: conceptual models, empirical challenges and interdiscipli- tion 111:1932–1937.
nary perspectives. Int J Epidemiol 31:285–293. Dabelea D, Knowler WC, Pettitt DJ. 2000. Effect of diabetes in pregnancy
Benyshek DC, Martin JF, Johnston CS. 2001. A reconsideration of the ori- on offspring: follow-up research in the Pima Indians. J Matern Fetal
gins of the type 2 diabetes epidemic among Native Americans and the Med 9:83–88.
implications for intervention policy. Med Anthropol 20:25–64. Dalziel SR, Parag V, Rodgers A, Harding JE. 2007. Cardiovascular risk fac-
Berger SL. 2007. The complex language of chromatin regulation during tors at age 30 following pre-term birth. Int J Epidemiol 36:907–915.
transcription. Nature 447:407–412. David R, Collins J Jr. 2007. Disparities in infant mortality: what’s genetics
Blane D, Berney L, Smith GD, Gunnell DJ, Holland P. 1999. Reconstruct- got to do with it? Am J Public Health 97:1191–1197.
ing the life course: health during early old age in a follow-up study based David RJ, Collins JW Jr. 1997. Differing birth weight among infants of
on the Boyd Orr Cohort. Public Health 113:117–124. U.S.-born blacks, African-born blacks, and U.S.-born whites. N Engl J
Bo S, Cavallo-Perin P, Scaglione L, Ciccone G, Pagano G. 2000. Low birth- Med 337:1209–1214.
weight and metabolic abnormalities in twins with increased susceptibil- Davidson K, Jonas BS, Dixon KE, Markovitz JH. 2000. Do depression
ity to Type 2 diabetes mellitus. Diabet Med 17:365–370. symptoms predict early hypertension incidence in young adults in the
Boas F. 1912. Changes in bodily form of descendants of immigrants. Am CARDIA study? Coronary artery risk development in young adults.
Anthropol 14:530–562. Arch Intern Med 160:1495–1500.
Bogdarina I, Welham S, King PJ, Burns SP, Clark AJ. 2007. Epigenetic Demerath EW, Choh AC, Czerwinski SA, Lee M, Sun SS, Chumlea WC,
modification of the renin-angiotensin system in the fetal programming Duren D, Sherwood RJ, Blangero J, Towne B, Siervogel RM. 2007.
of hypertension. Circ Res 100:520–526. Genetic and environmental influences on infant weight and weight
Bogin B. 1999. Patterns of human growth. Cambridge: Cambridge Univer- change: The Fels Longitudinal Study. Am J Hum Biol 19:692–702.
sity Press. Demissie K, Rhoads GG, Ananth CV, Alexander GR, Kramer MS, Kogan
Borders AE, Grobman WA, Amsden LB, Holl JL. 2007. Chronic stress and MD, Joseph KS. 2001. Trends in preterm birth and neonatal mortality
low birth weight neonates in a low-income population of women. Obstet among blacks and whites in the United States from 1989 to 1997. Am J
Gynecol 109(2, Part 1):331–338. Epidemiol 154:307–315.
Braveman PA, Cubbin C, Egerter S, Chideya S, Marchi KS, Metzler M, di Leonardo M. 2004. Human cultural diversity. Race and human varia-
Posner S. 2005. Socioeconomic status in health research: one size does tion: setting an agenda for future research and education. Alexandria,
not fit all. JAMA 294:2879–2888. VA: American Anthropological Association.
Bredy TW, Humpartzoomian RA, Cain DP, Meaney MJ. 2003. Partial re- Diorio J, Meaney MJ. 2007. Maternal programming of defensive responses
versal of the effect of maternal care on cognitive function through envi- through sustained effects on gene expression. J Psychiatry Neurosci
ronmental enrichment. Neuroscience 118:571–576. 32:275–284.
Brondolo E, Rieppi R, Kelly KP, Gerin W. 2003. Perceived racism and blood Dole N, Savitz DA, Hertz-Picciotto I, Siega-Riz AM, McMahon MJ, Buek-
pressure: a review of the literature and conceptual and methodological ens P. 2003. Maternal stress and preterm birth. Am J Epidemiol 157:14–
critique. Ann Behav Med 25:55–65. 24.
Brown R, Armelagos G. 2001. Apportionment of racial diversity: a review. Dole N, Savitz DA, Siega-Riz AM, Hertz-Picciotto I, McMahon MJ, Buek-
Evol Anthropol 10:34–40. ens P. 2004. Psychosocial factors and preterm birth among African
Burchard EG, Ziv E, Coyle N, Gomez SL, Tang H, Karter AJ, Mountain American and white women in central North Carolina. Am J Public
JL, Perez-Stable EJ, Sheppard D, Risch N. 2003. The importance of race Health 94:1358–1365.
and ethnic background in biomedical research and clinical practice. N Dominguez TP, Schetter CD, Mancuso R, Rini CM, Hobel C. 2005. Stress
Engl J Med 348:1170–1175. in African American pregnancies: testing the roles of various stress con-
Cabral H, Fried LE, Levenson S, Amaro H, Zuckerman B. 1990. Foreign- cepts in prediction of birth outcomes. Ann Behav Med 29:12–21.
born and US-born black women: differences in health behaviors and Donker GA, Labarthe DR, Harrist RB, Selwyn BJ, Wattigney W, Berenson
birth outcomes. Am J Public Health 80:70–72. GS. 1997. Low birth weight and blood pressure at age 7-11 years in a
Calafell F. 2003. Classifying humans. Nat Genet 33:435–436. biracial sample. Am J Epidemiol 145:387–397.
CDC. 2005. Racial/ethnic disparities in infant mortality—United States, Dörner G. 1975. Perinatal hormone levels and brain organization. In: Ana-
1995-2002. Morb Mortal Wkly Rep 54:553–556. tatomical neuroendocrinology. Basel: Karger. p 245–252.
CDC. 2007. National diabetes fact sheet: United States, 2007. Atlanta, GA: Drake AJ, Walker BR. 2004. The intergenerational effects of fetal program-
Centers for Disease Control and Prevention. ming: non-genomic mechanisms for the inheritance of low birth weight
Challis JR, Bloomfield FH, Bocking AD, Casciani V, Chisaka H, Connor K, and cardiovascular risk. J Endocrinol 180:1–16.
Dong X, Gluckman P, Harding JE, Johnstone J, Li W, Lye S, Okamura Drake AJ, Walker BR, Seckl JR. 2005. Intergenerational consequences of
K, Premyslova M. 2005. Fetal signals and parturition. J Obstet Gynaecol fetal programming by in utero exposure to glucocorticoids in rats. Am J
Res 31:492–499. Physiol Regul Integr Comp Physiol 288:R34–R38.
Clausson B, Cnattingius S, Axelsson O. 1998. Preterm and term births Dressler WW. 1991. Social class, skin color, and arterial blood pressure in
of small for gestational age infants: a population-based study of risk two societies. Ethn Dis 1:60–77.
factors among nulliparous women. Br J Obstet Gynaecol 105:1011–1017. Dressler WW, Bindon JR. 2000. The health consequences of cultural conso-
Collins JW Jr, David RJ, Handler A, Wall S, Andes S. 2004. Very low nance: Cultural dimensions of lifestyle, social support, and arterial blood
birthweight in African American infants: the role of maternal exposure to pressure in an African American community. Am Anthropol 102:244–
interpersonal racial discrimination. Am J Public Health 94:2132–2138. 260.
Collins JW Jr, Wu SY, David RJ. 2002. Differing intergenerational birth Dressler WW, Oths KS, Gravlee CC. 2005. Race and ethnicity in public
weights among the descendants of US-born and foreign-born whites and health research: models to explain health disparities. Annu Rev Anthro-
African Americans in Illinois. Am J Epidemiol 155:210–216. pol 34:231–252.
Cooper R, David R. 1986. The biological concept of race and its application Eriksson JG, Forsen T, Tuomilehto J, Jaddoe VW, Osmond C, Barker DJ.
to public health and epidemiology. J Health Polit Policy Law 11:97–116. 2002. Effects of size at birth and childhood growth on the insulin resist-
Cooper RS. 1993. Health and the social status of blacks in the United ance syndrome in elderly individuals. Diabetologia 45:342–348.
States. Ann Epidemiol 3:137–144. Fan Z, Lipsitz S, Egan B, Lackland D. 2000. The impact of birth weight
Cooper RS, Kaufman JS, Ward R. 2003. Race and genomics. N Engl J Med on the racial disparity of end-stage renal disease. Ann Epidemiol 10:
348:1166–1170. 459.
Cooper RS, Psaty BM. 2003. Genomics and medicine: distraction, incre- Farley TA, Mason K, Rice J, Habel JD, Scribner R, Cohen DA. 2006. The
mental progress, or the dawn of a new age? Ann Intern Med 138:576– relationship between the neighbourhood environment and adverse birth
580. outcomes. Paediatr Perinat Epidemiol 20:188–200.

American Journal of Human Biology

 EPIGENETICS AND THE EMBODIMENT OF RACE 13
Forna F, Jamieson DJ, Sanders D, Lindsay MK. 2003. Pregnancy outcomes Hirsch C, Anderson ML, Newman A, Kop W, Jackson S, Gottdiener J,
in foreign-born and US-born women. Int J Gynaecol Obstet 83:257–265. Tracy R, Fried LP. 2006. The association of race with frailty: the cardio-
Forrest CB, Riley AW. 2004. Childhood origins of adult health: a basis for vascular health study. Ann Epidemiol 16:545–553.
life-course health policy. Health Aff (Millwood) 23:155–164. Ho SM, Tang WY. 2007. Techniques used in studies of epigenome dysregu-
Forsdahl A. 1977. Are poor living conditions in childhood and adolescence lation due to aberrant DNA methylation: an emphasis on fetal-based
an important risk factor for arteriosclerotic heart disease? Br J Prev Soc adult diseases. Reprod Toxicol 23:267–282.
Med 31:91–95. Huxley R, Owen CG, Whincup PH, Cook DG, Rich-Edwards J, Smith GD,
Francis DD, Diorio J, Plotsky PM, Meaney MJ. 2002. Environmental Collins R. 2007. Is birth weight a risk factor for ischemic heart disease
enrichment reverses the effects of maternal separation on stress reactiv- in later life? Am J Clin Nutr 85:1244–1250.
ity. J Neurosci 22:7840–7843. IJzerman R, Stehouwer CD, de Geus EJ, van Weissenbruch MM, Dele-
Freathy RM, Weedon MN, Bennett A, Hypponen E, Relton CL, Knight B, marre-van de Waal HA, Boomsma DI. 2003. Low birth weight is associ-
Shields B, Parnell KS, Groves CJ, Ring SM, Pembrey ME, Ben-Shlomo ated with increased sympathetic activity: dependence on genetic factors.
Y, Strachan DP, Power C, Jarvelin MR, McCarthy MI, Davey Smith G, Circulation 108:566–571.
Hattersley AT, Frayling TM. 2007. Type 2 diabetes TCF7L2 risk geno- Iliadou A, Cnattingius S, Lichtenstein P. 2004. Low birthweight and type 2
types alter birth weight: a study of 24,053 individuals. Am J Hum Genet diabetes: a study on 11 162 Swedish twins. Int J Epidemiol 33:948–953;
80:1150–1161. discussion 953-954.
Geronimus AT. 2001. Understanding and eliminating racial inequalities in Jablonka E, Lamb MJ. 1995. Epigenetic inheritance and evolution: the
women’s health in the United States: the role of the weathering concep- Lamarckian dimension. Oxford, New York: Oxford University Press.
tual framework. J Am Med Womens Assoc 56:133–136;149–150. 346 p.
Geronimus AT, Hicken M, Keene D, Bound J. 2006. ‘‘Weathering’’ and age Jensen CB, Storgaard H, Madsbad S, Richter EA, Vaag AA. 2007. Altered
patterns of allostatic load scores among blacks and whites in the United skeletal muscle fiber composition and size precede whole-body insulin
States. Am J Public Health 96:826–833. resistance in young men with low birth weight. J Clin Endocrinol Metab
Giscombe CL, Lobel M. 2005. Explaining disproportionately high rates of 92:1530–1534.
adverse birth outcomes among African Americans: the impact of stress, Jenuwein T, Allis CD. 2001. Translating the histone code. Science
racism, and related factors in pregnancy. Psychol Bull 131:662–683. 293:1074–1080.
Gluckman PD, Hanson MA. 2004. Maternal constraint of fetal growth and Jirtle RL, Skinner MK. 2007. Environmental epigenomics and disease sus-
its consequences. Semin Fetal Neonatal Med 9:419–425. ceptibility. Nat Rev Genet 8:253–262.
Gluckman PD, Hanson MA. 2005. Metabolic disease: evolutionary, devel- Jonas BS, Lando JF. 2000. Negative affect as a prospective risk factor for
opmental and transgenerational influences. Nestle Nutr Workshop Ser hypertension. Psychosom Med 62:188–196.
Pediatr Program 55:17–27. Jorde L, Wooding S. 2004. Genetic variation, classification, and ‘‘race’’. Nat
Gluckman PD, Hanson MA. 2006. Developmental origins of health and Genet 36:S28–S33.
disease. Cambridge: Cambridge University Press. Kaplan GA, Lynch JW. 2001. Is economic policy health policy? Am J Public
Gluckman PD, Hanson MA, Beedle AS. 2007a. Non-genomic transgenera- Health 91:351–353.
tional inheritance of disease risk. Bioessays 29:145–154. Kaufman JS, Cooper RS. 1999. Seeking causal explanations in social epi-
Gluckman PD, Hanson MA, Cooper C, Thornburg KL. 2008. Effect of in demiology. Am J Epidemiol 150:113–120.
utero and early-life conditions on adult health and disease. New Engl J Kennedy D. 2002. Breakthrough of the year. Science 298:2283.
Med 359:61–73. Keppel KG, Pearcy JN, Wagener DK. 2002. Trends in racial and ethnic-
Gluckman PD, Lillycrop KA, Vickers MH, Pleasants AB, Phillips ES, Bee- specific rates for the health status indicators: United States, 1990-98.
dle AS, Burdge GC, Hanson MA. 2007b. Metabolic plasticity during Healthy People 2000 Stat Notes 23:1–16.
mammalian development is directionally dependent on early nutritional Kermack W, McKendrick A, McKinlay P. 1934. Death rates in Great Brit-
status. Proc Natl Acad Sci USA 104:12796–12800. ain and Sweden: some general regularities and their significance. Lan-
Goodman AH. 2000. Why genes don’t count (for racial differences in cet 226:698–703.
health). Am J Pub Health 90:1699–1702. Kistka ZA, Palomar L, Lee KA, Boslaugh SE, Wangler MF, Cole FS,
Gordon MM. 1964. Assimilation in American life: the role of race, religion DeBaun MR, Muglia LJ. 2007. Racial disparity in the frequency of recur-
and national origins. New York: University of Massachusetts Press. rence of preterm birth. Am J Obstet Gynecol 196:131e1–131e6.
Grady SC. 2006. Racial disparities in low birthweight and the contribution Kittles RA, Weiss KM. 2003. Race, ancestry, and genes: Implications for
of residential segregation: a multilevel analysis. Soc Sci Med 63:3013– defining disease risk. Annu Rev Genomics Hum Genet 4:33–67.
3029. Kleinman JC, Fingerhut LA, Prager K. 1991. Differences in infant mortal-
Graham H, Power C. 2004. Childhood disadvantage and health inequal- ity by race, nativity status, and other maternal characteristics. Am J Dis
ities: a framework for policy based on lifecourse research. Child Care Child 145:194–199.
Health Dev 30:671–678. Kramer MS, Ananth CV, Platt RW, Joseph KS. 2006. US black vs white dis-
Graham J, Zhang L, Schwalberg R. 2007. Association of maternal chronic parities in foetal growth: physiological or pathological? Int J Epidemiol
disease and negative birth outcomes in a non-Hispanic Black-White Mis- 35:1187–1195.
sissippi birth cohort. Public Health Nurs 24:311–317. Kramer MS, Joseph KS. 1996. Enigma of fetal/infant-origins hypothesis.
Gravlee CC, Dressler WW, Bernard HR. 2005. Skin color, social classifica- Lancet 348:1254–1255.
tion, and blood pressure in southeastern Puerto Rico. Am J Public Krieger N. 2000. Refiguring ‘‘race’’: epidemiology, racialized biology,
Health 95:2191–2197. and biological expressions of race relations. Int J Health Serv 30:211–
Gravlee CC, Sweet E. 2008. Race, ethnicity, and racism in medical anthro- 216.
pology, 1977-2002. Med Anthropol Q 22:27–51. Krieger N. 2005. Stormy weather: race, gene expression, and the science of
Grewal SI, Elgin SC. 2007. Transcription and RNA interference in the for- health disparities. Am J Public Health 95:2155–2160.
mation of heterochromatin. Nature 447:399–406. Krieger N, Davey Smith G. 2004. ‘‘Bodies count,’’ and body counts: social
Gunnar MR. 1998. Quality of early care and buffering of neuroendocrine epidemiology and embodying inequality. Epidemiol Rev 26:92–103.
stress reactions: potential effects on the developing human brain. Prev Kuh D, Ben-Shlomo Y, Lynch J, Hallqvist J, Power C. 2003. Life course epi-
Med 27:208–211. demiology. J Epidemiol Community Health 57:778–783.
Gupta M, Gupta R, Pareek A, Bhatia R, Kaul V. 2007. Low birth weight Kuh D, Shlomo YB. 2004. A life course approach to chronic disease epide-
and insulin resistance in mid and late childhood. Indian Pediatr 44:177– miology. Oxford: Oxford University Press.
184. Kuzawa CW. 2005. The fetal origins of developmental plasticity: are fetal
Halfon N, Hochstein M. 2002. Life course health development: an inte- cues reliable predictors of future nutritional environments? Am J Hum
grated framework for developing health, policy, and research. Milbank Biol 175–21.
Q 80:433–479, iii. Kuzawa CW. 2008. The developmental origins of adult health: intergenera-
Harrell JP, Hall S, Taliaferro J. 2003. Physiological responses to racism tional inertia in adaptation and disease. In: Trevathan W, Smith EO,
and discrimination: an assessment of the evidence. Am J Public Health McKenna JJ, editors. Evolution and health. Oxford: Oxford University
93:243–248. Press. pp 325–349.
Harrison FV. 1995. The persistent power of ‘‘race’’ in the cultural and polit- Kuzawa CW, Adair LS. 2003. Lipid profiles in adolescent Filipinos: relation
ical economy of racism. Annu Rev Anthropol 24:47–74. to birth weight and maternal energy status during pregnancy. Am J Clin
Hattersley AT, Tooke JE. 1999. The fetal insulin hypothesis: an alternative Nutr 77960–966.
explanation of the association of low birthweight with diabetes and vas- Kuzawa CW, Adair LS. 2004. A supply-demand model of fetal energy suffi-
cular disease. Lancet 353:1789–1792. ciency predicts lipid profiles in male but not female Filipino adolescents.
Hinds DA, Stuve LL, Nilsen GB, Halperin E, Eskin E, Ballinger DG, Eur J Clin Nutr 58:438–448.
Frazer KA, Cox DR. 2005. Whole-genome patterns of common DNA vari- Kuzawa CW, Pike IL. 2005. Introduction. Fetal origins of developmental
ation in three human populations. Science 307:1072–1079. plasticity. Am J Hum Biol 17:1–4.

American Journal of Human Biology

14 C.W. KUZAWA AND E. SWEET

Lampl M, Jeanty P. 2004. Exposure to maternal diabetes is associated with McDade TW, Hawkley LC, Cacioppo JT. 2006. Psychosocial and behav-
altered fetal growth patterns: a hypothesis regarding metabolic alloca- ioral predictors of inflammation in middle-aged and older adults: The
tion to growth under hyperglycemic-hypoxemic conditions. Am J Hum Chicago Health, Aging, and Social Relations Study. Psychosom Med
Biol 16:237–263. 68: 376–381.
Lampl M, Kuzawa CW, Jeanty P. 2002. Infants thinner at birth exhibit McEwen BS. 2001. From molecules to mind. Stress, individual differences,
smaller kidneys for their size late in gestation in a sample of fetuses and the social environment. Ann N Y Acad Sci 935:42–49.
with appropriate growth. Am J Hum Biol 14:398–406. McMillen IC, Robinson JS. 2005. Developmental origins of the metabolic
Langley-Evans SC, Langley-Evans AJ, Marchand MC. 2003. Nutritional syndrome: prediction, plasticity, and programming. Physiol Rev 85:571–
programming of blood pressure and renal morphology. Arch Physiol Bio- 633.
chem 111:8–16. Mensah GA, Mokdad AH, Ford ES, Greenlund KJ, Croft JB. 2005. State of
Lao O, van Duijn K, Kersbergen P, de Knijff P, Kayser M. 2006. Proportion- disparities in cardiovascular health in the United States. Circulation
ing whole-genome single-nucleotide-polymorphism diversity for the 111:1233–1241.
identification of geographic population structure and genetic ancestry. Miura K, Nakagawa H, Tabata M, Morikawa Y, Nishijo M, Kagamimori S.
Am J Hum Genet 78:680–690. 2001. Birth weight, childhood growth, and cardiovascular disease risk
LaVeist T. 2005. Disentangling race and socioeconomic status: a key to factors in Japanese aged 20 years. Am J Epidemiol 153:783–789.
understanding health inequalities. J Urban Health 82(2 Suppl 3):iii26– Morton S. 2006. Maternal nutrition and fetal growth and development. In:
iii34. Gluckman P, Hanson M, editors. Developmental origins of health and
Laviola G, Rea M, Morley-Fletcher S, Di Carlo S, Bacosi A, De Simone R, disease. Cambridge: CUP.
Bertini M, Pacifici R. 2004. Beneficial effects of enriched environment on Mustillo S, Krieger N, Gunderson EP, Sidney S, McCreath H, Kiefe CI.
adolescent rats from stressed pregnancies. Eur J Neurosci 20:1655–1664. 2004. Self-reported experiences of racial discrimination and black-white
Law CM, Egger P, Dada O, Delgado H, Kylberg E, Lavin P, Tang GH, von differences in preterm and low-birthweight deliveries: The CARDIA
Hertzen H, Shiell AW, Barker DJ. 2001. Body size at birth and blood Study. Am J Public Health 94:2125–2131.
pressure among children in developing countries. Int J Epidemiol Mzayek F, Hassig S, Sherwin R, Hughes J, Chen W, Srinivasan S, Beren-
30:52–57. son G. 2007. The association of birth weight with developmental trends
Lawlor DA, Riddoch CJ, Page AS, Andersen LB, Wedderkopp N, Harro M, in blood pressure from childhood through mid-adulthood: The Bogalusa
Stansbie D, Smith GD. 2005. Infant feeding and components of the met- Heart study. Am J Epidemiol 166:413–420.
abolic syndrome: findings from the European Youth Heart Study. Arch Mzayek F, Sherwin R, Fonseca V, Valdez R, Srinivasan SR, Cruickshank
Dis Child 90:582–588. JK, Berenson GS. 2004. Differential association of birth weight with car-
Leary S, Davey Smith G, Ness A. 2006. Smoking during pregnancy and diovascular risk variables in African-Americans and whites: The Boga-
components of stature in offspring. Am J Hum Biol 18:502–512. lusa Heart Study. Ann Epidemiol 14:258–264.
Leon DA, Lithell HO, Vagero D, Koupilova I, Mohsen R, Berglund L, Norbeck JS, DeJoseph JF, Smith RT. 1996. A randomized trial of an empir-
Lithell UB, McKeigue PM. 1998. Reduced fetal growth rate and ically-derived social support intervention to prevent low birthweight
increased risk of death from ischaemic heart disease: cohort study of among African American women. Soc Sci Med 43:947–954.
15000 Swedish men and women born 1915-29. BMJ 317:241–245. Oberg S, Ge D, Cnattingius S, Svensson A, Treiber FA, Snieder H, Iliadou
Levitt NS, Lambert EV, Woods D, Hales CN, Andrew R, Seckl JR. 2000. A. 2007. Ethnic differences in the association of birth weight and blood
Impaired glucose tolerance and elevated blood pressure in low birth pressure: The Georgia Cardiovascular Twin Study. Am J Hypertens
weight, nonobese, young south african adults: early programming of cor- 20:1235–1241.
tisol axis. J Clin Endocrinol Metab 85:4611–4618. Oken E, Gillman MW. 2003. Fetal origins of obesity. Obes Res 11:496–506.
Lewontin R. 1972. The apportionment of human diversity. Evol Biol 6:381– Oken E, Levitan EB, Gillman MW. 2008. Maternal smoking during preg-
398. nancy and child overweight: systematic review and meta-analysis. Int J
Li C, Huang TK, Cruz ML, Goran MI. 2006. Birth weight, puberty, and Obes 32:201–210.
systolic blood pressure in children and adolescents: a longitudinal analy- Ong KK. 2006. Size at birth, postnatal growth and risk of obesity. Horm
sis. J Hum Hypertens 20:444–450. Res 65 (Suppl 3):65–69.
Li C, Johnson MS, Goran MI. 2001. Effects of low birth weight on insulin Orr ST, James SA, Miller CA, Barakat B, Daikoku N, Pupkin M, Engstrom
resistance syndrome in caucasian and African-American children. Dia- K, Huggins G. 1996. Psychosocial stressors and low birthweight in an
betes Care 24:2035–2042. urban population. Am J Prev Med 12:459–466.
Lillie-Blanton M, Laveist T. 1996. Race/ethnicity, the social environment, Oths KS, Dunn LL, Palmer NS. 2001. A prospective study of psychosocial
and health. Soc Sci Med 43:83–91. job strain and birth outcomes. Epidemiology 12:744–746.
Lillycrop KA, Phillips ES, Jackson AA, Hanson MA, Burdge GC. 2005. Die- Otten MW Jr, Teutsch SM, Williamson DF, Marks JS. 1990. The effect of
tary protein restriction of pregnant rats induces and folic acid supple- known risk factors on the excess mortality of black adults in the United
mentation prevents epigenetic modification of hepatic gene expression States. JAMA 263:845–850.
in the offspring. J Nutr 135:1382–1386. Ounsted M, Ounsted C. 1968. Rate of intra-uterine growth. Nature
Lillycrop KA, Slater-Jefferies JL, Hanson MA, Godfrey KM, Jackson AA, 220:599–600.
Burdge GC. 2007. Induction of altered epigenetic regulation of the he- Ounsted M, Scott A, Ounsted C. 1986. Transmission through the female
patic glucocorticoid receptor in the offspring of rats fed a protein-re- line of a mechanism constraining human fetal growth. Ann Hum Biol
stricted diet during pregnancy suggests that reduced DNA methyltrans- 13:143–151.
ferase-1 expression is involved in impaired DNA methylation and Pager D. 2003. The mark of a criminal record. Am J Soc 108:937–975.
changes in histone modifications. Br J Nutr 97:1064–1073. Palloni A. 2006. Reproducing inequalities: luck, wallets, and the enduring
Lu Q, Lu MC, Schetter CD. 2005. Learning from success and failure in psy- effects of childhood health. Demography 43:587–615.
chosocial intervention: an evaluation of low birth weight prevention tri- Paneth N, Ahmed F, Stein A. 1996. Early nutritional origins of hyperten-
als. J Health Psychol 10:185–195. sion: a hypothesis still lacking support. J Hypertens 14 (Suppl 5):s121–
Lucas A. 1991. Programming by early nutrition in man. Ciba Found Symp s129.
156:38–50; discussion 50–55. Pappas G, Queen S, Hadden W, Fisher G. 1993. The increasing disparity in
Lumey LH. 1992. Decreased birthweights in infants after maternal in mortality between socioeconomic groups in the United States, 1960 and
utero exposure to the Dutch famine of 1944–1945. Paediatr Perinat Epi- 1986. N Engl J Med 329:103–109.
demiol 6:240–253. Pettitt DJ, Aleck KA, Baird HR, Carraher MJ, Bennett PH, Knowler WC.
Lunde A, Melve KK, Gjessing HK, Skjaerven R, Irgens LM. 2007. Genetic 1988. Congenital susceptibility to NIDDM. Role of intrauterine environ-
and environmental influences on birth weight, birth length, head cir- ment. Diabetes 37:622–628.
cumference, and gestational age by use of population-based parent-off- Pettitt DJ, Knowler WC, Bennett PH, Aleck KA, Baird HR. 1987. Obesity
spring data. Am J Epidemiol 165:734–741. in offspring of diabetic Pima Indian women despite normal birth weight.
Luyckx VA, Brenner BM. 2005. Low birth weight, nephron number, and Diabetes Care 10:76–80.
kidney disease. Kidney Int Suppl 97:S68–S77. Pico C, Oliver P, Sanchez J, Miralles O, Caimari A, Priego T, Palou A.
Mackey MC, Boyle JS. 2000. An explanatory model of preterm labor. 2007. The intake of physiological doses of leptin during lactation in rats
J Transcult Nurs 11:254–263. prevents obesity in later life. Int J Obes (Lond) 31:1199–1209.
McCord C, Freeman HP. 1990. Excess mortality in Harlem. N Engl J Med Pike IL. 2005. Maternal stress and fetal responses: evolutionary perspec-
322:173–177. tives on preterm delivery. Am J Hum Biol 17:55–65.
McCormick CM, Mathews IZ. 2007. HPA function in adolescence: role of Pollitt RA, Rose KM, Kaufman JS. 2005. Evaluating the evidence for mod-
sex hormones in its regulation and the enduring consequences of expo- els of life course socioeconomic factors and cardiovascular outcomes: a
sure to stressors. Pharmacol Biochem Behav 86:220–233. systematic review. BMC Public Health 5:7.
McCowan LM, Buist RG, North RA, Gamble G. 1996. Perinatal morbidity Prentice AM, Moore SE. 2005. Early programming of adult diseases in
in chronic hypertension. Br J Obstet Gynaecol 103:123–129. resource poor countries. Arch Dis Child 90:429–432.

American Journal of Human Biology

 EPIGENETICS AND THE EMBODIMENT OF RACE 15
Ramakrishnan U, Martorell R, Schroeder DG, Flores R. 1999. Role of Suzuki MM, Bird A. 2008. DNA methylation landscapes: provocative
intergenerational effects on linear growth. J Nutr 129(2S Suppl):544S– insights from epigenomics. Nat Rev Genet 9:465–476.
549S. Sweet E, McDade TW, Kiefe CI, Liu K. 2007. Relationships between skin
Ray JG, Vermeulen MJ, Shapiro JL, Kenshole AB. 2001. Maternal and color, income, and blood pressure among African Americans in the CAR-
neonatal outcomes in pregestational and gestational diabetes mellitus, DIA Study. Am J Public Health 97:2253–2259.
and the influence of maternal obesity and weight gain: The DEPOSIT Symonds ME, Gardner DS. 2006. Experimental evidence for early nutri-
Study. Diabetes Endocrine Pregnancy Outcome Study in Toronto. QJM tional programming of later health in animals. Curr Opin Clin Nutr
94:347–356. Metab Care 9:278–283.
Reagan PB, Salsberry PJ. 2005. Race and ethnic differences in determi- Tang H, Quertermous T, Rodriguez B, Kardia SL, Zhu X, Brown A, Pan-
nants of preterm birth in the USA: broadening the social context. Soc Sci kow JS, Province MA, Hunt SC, Boerwinkle E, Schork NJ, Risch NJ.
Med 60:2217–2228. 2005. Genetic structure, self-identified race/ethnicity, and confounding
Redon R, Ishikawa S, Fitch KR, Feuk L, Perry GH, Andrews TD, Fiegler in case-control association studies. Am J Hum Genet 76:268–275.
H, Shapero MH, Carson AR, Chen W, Cho EK, Dallaire S, Freeman JL, Tian JY, Cheng Q, Song XM, Li G, Jiang GX, Gu YY, Luo M. 2006. Birth
Gonzalez JR, Gratacos M, Huang J, Kalaitzopoulos D, Komura D, Mac- weight and risk of type 2 diabetes, abdominal obesity and hypertension
Donald JR, Marshall CR, Mei R, Montgomery L, Nishimura K, Okamura among Chinese adults. Eur J Endocrinol 155:601–607.
K, Shen F, Somerville MJ, Tchinda J, Valsesia A, Woodwark C, Yang F, Troxel WM, Matthews KA, Bromberger JT, Sutton-Tyrrell K. 2003.
Zhang J, Zerial T, Zhang J, Armengol L, Conrad DF, Estivill X, Tyler- Chronic stress burden, discrimination, and subclinical carotid artery
Smith C, Carter NP, Aburatani H, Lee C, Jones KW, Scherer SW, Hurles disease in African American and Caucasian women. Health Psychol
ME. 2006. Global variation in copy number in the human genome. Na- 22:300–309.
ture 444:444–454. Vickers MH, Gluckman PD, Coveny AH, Hofman PL, Cutfield WS, Gertler
Reik W. 2007. Stability and flexibility of epigenetic gene regulation in A, Breier BH, Harris M. 2005. Neonatal leptin treatment reverses devel-
mammalian development. Nature 447:425–432. opmental programming. Endocrinology 146:4211–4216.
Relethford J. 2002. Apportionment of global human genetic diversity based Vlietinck R, Derom R, Neale MC, Maes H, van Loon H, Derom C, Thiery
on craniometrics and skin color. Am J Phys Anthropol 118:393–398. M. 1989. Genetic and environmental variation in the birth weight of
Risch N, Burchard E, Ziv E, Tang H. 2002. Categorization of humans twins. Behav Genet 19:151–161.
in biomedical research: genes, race and disease. Genome Biol Waddington C. 1942. The epigenotype. Endeavour 1:18–20.
3:comment2007. Waterland RA, Jirtle RL. 2004. Early nutrition, epigenetic changes at
Romeo RD, McEwen BS. 2006. Stress and the adolescent brain. Ann N Y transposons and imprinted genes, and enhanced susceptibility to adult
Acad Sci 1094:202–214. chronic diseases. Nutrition 20:63–68.
Rosenberg NA, Pritchard JK, Weber JL, Cann HM, Kidd KK, Zhivotovsky Waterland RA, Michels KB. 2007. Epigenetic epidemiology of the develop-
LA, Feldman MW. 2002. Genetic structure of human populations. Sci- mental origins hypothesis. Annu Rev Nutr 27:363–388.
ence 298:2381–2385. Weaver IC, Cervoni N, Champagne FA, D’Alessio AC, Sharma S, Seckl JR,
Sandman CA, Wadhwa PD, Chicz-DeMet A, Dunkel-Schetter C, Porto M. Dymov S, Szyf M, Meaney MJ. 2004. Epigenetic programming by mater-
1997. Maternal stress, HPA activity, and fetal/infant outcome. Ann N Y nal behavior. Nat Neurosci 7:847–854.
Acad Sci 814:266–275. Weismantel M. 1997. White cannibals: fantasies of racial violence in the
Sarich V, Miele F. 2004. Race: the reality of human differences. Boulder: Andes. Identities 4:9–43.
Westview Press. West-Eberhard MJ. 2003. Developmental plasticity and evolution. Oxford:
Saunders E. 1995. Hypertension in minorities: Blacks. Am J Hypertens Oxford University Press..
8(12, Part 2):115s–119s. Whitfield JB, Treloar SA, Zhu G, Martin NG. 2001. Genetic and non-
Seckl JR, Meaney MJ. 2004. Glucocorticoid programming. Ann N Y Acad genetic factors affecting birth-weight and adult Body Mass Index. Twin
Sci 1032:63–84. Res 4:365–370.
Serre D, Paabo S. 2004. Evidence for gradients of human genetic diversity Wightkin J, Magnus JH, Farley TA, Boris NW, Kotelchuck M. 2007. Psy-
within and among continents. Genome Res 14:1679–1685. chosocial predictors of being an underweight infant differ by racial
Shanklin E. 1994. Anthropology and race. Belmont, CA: Wadsworth. group: a prospective study of Louisiana WIC program participants.
Shapiro TM. 2004. The hidden cost of being African American: how wealth Matern Child Health J 11:49–55.
perpetuates inequality. Oxford: Oxford University Press. Williams DR. 1999. Race, socioeconomic status, and health. The added
Silverman BL, Metzger BE, Cho NH, Loeb CA. 1995. Impaired glucose tol- effects of racism and discrimination. Ann N Y Acad Sci 896:173–188.
erance in adolescent offspring of diabetic mothers. Relationship to fetal Williams DR. 2005. The health of U.S. racial and ethnic populations. J Ger-
hyperinsulinism. Diabetes Care 18:611–617. ontol B Psychol Sci Soc Sci 60:53–62.
Sinclair KD, Lea RG, Rees WD, Young LE. 2007. The developmental ori- Williams DR, Collins C. 1995. US socioeconomic and racial differences in
gins of health and disease: current theories and epigenetic mechanisms. health: Patterns and explanations. Annu Rev Sociol 21:349–386.
In: Juengel JL, Murray JF, Smith MF, editors. Reproduction in domes- Williams DR, Jackson PB. 2005. Social sources of racial disparities in
tic ruminants VI. Nottingham: Nottingham University Press. p 425– health. Health Aff (Millwood) 24:325–334.
444. Williams DR, Neighbors H. 2001. Racism, discrimination and hyperten-
Singh GK, Yu SM. 1996. Adverse pregnancy outcomes: Differences sion: evidence and needed research. Ethn Dis 11:800–816.
between US- and foreign-born women in major US racial and ethnic Worthman CM, Kuzara J. 2005. Life history and the early origins of health
groups. Am J Public Health 86:837–843. differentials. Am J Hum Biol 17:95–112.
Smith GD. 2000. Learning to live with complexity: ethnicity, socioeconomic Wyatt SB, Williams DR, Calvin R, Henderson FC, Walker ER, Winters K.
position, and health in Britain and the United States. Am J Public 2003. Racism and cardiovascular disease in African Americans. Am J
Health 90:1694–1698. Med Sci 325:315–331.
Smith GD. 2003. Health inequalities: lifecourse approaches. Bristol: The Yajnik CS. 2004. Early life origins of insulin resistance and type 2 diabetes
Policy Press. in India and other Asian countries. J Nutr 134:205–210.
Stern MP, Bartley M, Duggirala R, Bradshaw B. 2000. Birth weight and Yehuda R, Bierer LM. 2008. Transgenerational transmission of cortisol
the metabolic syndrome: thrifty phenotype or thrifty genotype? Diabetes and PTSD risk. Prog Brain Res 167:121–135.
Metab Res Rev 16:88–93. Zhang T, Guan H, Arany E, Hill DJ, Yang K. 2007. Maternal protein
Stöger R. 2008. The thrifty epigenotype: an acquired and inherited predis- restriction permanently programs adipocyte growth and development in
position for obesity and diabetes? Bioessays 30:156–166. adult male rat offspring. J Cell Biochem 101:381–388.

American Journal of Human Biology